EXCEED THE SPACE PROVIDED. Microbes responsible for infectious disease outbreaks - from either natural, e.g., multidrug-resistant Staphylococcus aureus, or bioterrorism, e.g.,anthrax, causes - are typically virulent forms of a single species. Population genomics, which is the study of intraspecific variations of a single microbial species, is therefore a key to uncovering virulence factors and disease-causing mechanisms, as well as tracking the sources of microbial disease. The Evolutionary Informatics Lab at Hunter College is a vital partner in a multi- institutional team of comparative genomics of Borrelia burgdorferi, the Lyme disease agent. We are responsible for the bioinformatics system development and evolutionary statistical analysis, including experimental design, data processing and warehousing, and evolutionary analysis. Through evolutionary statistical analysis our lab discovered a surprisingly high amount of genetic exchange and plasmid transfer in B. burgdorferi. To continue our success in characterizing the population structure of pathogenic bacteria based on comparative genomics, I propose a 5-years project to develop a bioinformatics infrastructure dedicated to the analysis of population variations of pathogenic microbes. We have two specific aims in this proposal. First, we will develop a novel statistical framework - non-housekeeping multilocus sequencing typing (NH-MLST) - of analyzing the bacterial intraspecific variations in a phylogenetic context, with the goal of detecting genes responsible for adaptive evolution in bacteria. Second, we will develop a computational infrastructure dedicated to the comparative genomics of closely related bacteria. As comparative microbial sequencing becomes a standard approach for discovering virulence factors and tracing microbial outbreaks, the proposed project will provide a robust evolutionary analytical framework and an efficient computational system for comparative microbial genomics.